We have developed a powerful mouse model to study visual circuit development in the dorsal lateral geniculate nucleus (dLGN), the thalamic relay of retinal information to the visual cortex. To date, studies have focused largely on the retinogeniculate pathway and the refinement of connections between retinal ganglion cells and dLGN neurons. While many of the changes in retinogeniculate axon patterning and connectivity have been delineated, information about two fundamental aspects of organization is still lacking. One relates to the cell- class specificity of connections between RGCs and dLGN cells and the other pertains to the source of connections to dLGN cells, namely the wealth of input that arises from nonretinal projections. Both aspects are critical for a contemporary understanding of thalamic function, which is to provide a faithful relay of retinal signals but at the same time to modulate the gain of transmission based on changes in behavioral state. The goals of this proposal are to assess how retinogeniculate refinement relates to class specific connections, whether retinogeniculate refinement proceeds in a way that establishes separable parallel visual channels, to determine when nonretinal projections develop in dLGN, and to assess how such input coordinates with the innervation and refinement of the retinogeniculate pathway. To address these issues, we will conduct anatomical, electrophysiological, and optogenetic experiments in transgenic mice that allow for the visualization and experimental isolation of specific cell types arising from the retina or from such nonretinal sources as layer VI of visual cortex, the thalamic reticular nucleus, and cholinergic nuclei of the brainstem. These studies will provide valuable information about how the developing brain forms precise patterns of connections and offer further insight into the study and treatment of developmental and neurological disorders that result from the formation of abnormal patterns of connectivity.